Every trading decision starts with information. What you do with that information — hold it, act on it, dismiss it — depends heavily on whether you first asked a more basic question: where did this come from, and what does the source want? Most newer traders skip that question entirely. They evaluate the claim. They almost never evaluate the claimer.

Source hygiene is the discipline of examining an information source's credibility, incentive structure, and evidence tier before the claim inside it gets any weight in your thinking. It is not about being cynical about every piece of data you encounter. It is about calibrating how much a given piece of information should actually move your reasoning — and recognizing when it should move it not at all.

Why This Matters More Than the Information Itself

Markets produce a continuous stream of commentary, analysis, screenshots, stories, and statistics. Much of it is packaged to look like evidence. Very little of it meets the standard of evidence. The gap between those two things — between looking like evidence and being evidence — is where a substantial portion of preventable trading errors originate.

When you act on weak information that looked strong, you are not making a decision based on the market. You are making a decision based on how confident someone else sounded. Those are entirely different inputs, and confusing them is a process failure before it is ever a position failure.

Source hygiene is the skill that sits upstream of every other analytical skill. A well-constructed thesis built on unreliable inputs is still a flawed thesis. Getting the inputs right first is the more efficient path.

The Evidence Tier Model

The most practical mental model for source evaluation is a simple hierarchy: primary data, secondary commentary, and hearsay. Each tier has a different relationship to verifiable reality, and each deserves a proportionally different amount of trust.

Tier 1 — Primary Data

Primary data is information that comes directly from the source of the event, without being filtered, interpreted, or repackaged by a third party. In a market context, this includes regulatory filings, official earnings releases, audited financial statements, exchange-published price and volume records, and central bank policy announcements. The defining feature of primary data is traceability: you can follow the chain back to the originating document and verify the number yourself.

Primary data can still be wrong — filings contain errors, auditors miss things, definitions change — but when it is wrong, the wrongness is knowable and correctable. The record exists. That traceability is what makes primary data the highest-value input.

Tier 2 — Secondary Commentary

Secondary commentary is an interpretation of primary data produced by someone other than the originating entity. Analyst reports, research summaries, earnings recaps, and explanatory articles all fall here. Secondary sources add context and perspective, which has genuine value. They also introduce the author's framing, selective emphasis, and — crucially — the author's incentives. A secondary source is always a layer of interpretation between you and the underlying fact. That layer can clarify, but it can also distort.

The right question for any secondary source is not just "is this accurate?" but "what does this author or publisher gain if I believe this?"

Tier 3 — Hearsay and Social Evidence

Hearsay is information passed between people without a traceable link to a primary source. It includes anonymous screenshots showing large returns, forum posts citing unnamed insiders, and second-hand summaries of things someone else claimed to have read. Hearsay can occasionally be directionally true. It is structurally unverifiable, which means you cannot distinguish the true cases from the false ones without additional evidence — at which point you are relying on that additional evidence, not the hearsay.

Social proof is a specific and powerful form of hearsay pressure. When many people confidently repeat the same unverified claim, the volume of repetition creates a feeling of reliability that the underlying evidence does not support. Recognizing this pattern — many voices, single untraced origin — is a core source hygiene skill.

The Incentive Filter

Tier classification alone is not enough. A primary-tier document can still be produced by someone with a strong incentive to present it in a particular light. The incentive filter asks a separate question for every source: what does this source gain if I update my beliefs in the direction they are pointing?

A person sharing a screenshot of a large gain has an incentive to share that screenshot specifically because it is a large gain. The population of their trading decisions — including the losses — is not visible. A company filing an investor presentation has an incentive to frame forward guidance optimistically. An analyst whose firm has existing positions in a sector has a structural interest in maintaining or expanding attention to that sector.

None of these incentives mean the information is wrong. They mean the information is tilted, and the tilt is predictable from the incentive. Adjusting for a predictable tilt is a basic epistemic skill. Ignoring the tilt and treating the information as neutral is a reliable way to absorb someone else's bias as your own conviction.

The Survivorship Problem in Success Stories

One specific incentive pattern deserves its own attention because it is so pervasive in trading content: survivorship in success narratives. When you see a story about a large winning trade, you are seeing a story that was worth telling. The stories not worth telling — the partial losses, the full losses, the strategies that worked for six months and then stopped — do not circulate with the same energy.

This is not deception in most cases. People share what worked. The problem is that the sample of stories that reaches you is not a representative sample of all outcomes — it is a sample filtered for success. Drawing conclusions about a strategy or approach from that filtered sample produces systematically overconfident conclusions. The question to ask when encountering any success story is: what happened to the people who tried this and are not telling me about it?

A Five-Point Vetting Method

Before any piece of information informs a decision in the simulator or in your thinking about markets, run it through these five questions in order:

  1. Who said it? Name the actual source — not "I read that" or "people are saying," but the specific entity or document. If you cannot name it, you are operating on hearsay.
  2. What is their incentive? Ask what this source gains if you act on this information. Be specific. "They want to inform people" is not an incentive analysis. "They benefit if more people hold this asset" is.
  3. Is this primary data or an interpretation of primary data? If it is an interpretation, can you trace it back to the underlying primary source? If no primary source is cited or findable, classify it as Tier 3 until proven otherwise.
  4. Can you verify it independently? A claim that cannot be checked from a second independent primary source should carry proportionally less weight. Convergent evidence from multiple independent primary sources is far more reliable than a confident single source.
  5. What is missing? What would a complete picture include that this source does not show? What counter-evidence would weaken this claim, and why might the source have omitted it?

A Hypothetical Example: The Anonymous Win Screenshot

Imagine you encounter a screenshot posted in an online trading community. It shows a brokerage account dashboard with a large percentage gain on a single position over several weeks. The post is anonymous. No entry context is provided, no strategy explanation, no information about the size of the account before the trade, and no mention of other trades taken during the same period.

Run the five-point vetting method. Who said it? Anonymous — Tier 3 immediately. What is their incentive? At minimum, social recognition for having made a large gain. Is it primary data? The screenshot shows a number on a screen. It is not traceable to a verified transaction record. Can you verify it? No. What is missing? The rest of the account's history, the strategy logic, whether this was one of ten simultaneous positions, whether the account existed before this trade.

The screenshot is not evidence of a reliable approach. It is evidence that one person claims to have had one large gain. Those are not the same thing. Acting as though they are is not a market error — it is a source error that precedes and causes market errors.

Practicing This in the Abu Simulator

Abu Terminal is built on verified historical price data — exchange-sourced, publicly traceable records. Every number displayed in a Speed Run scenario can, in principle, be traced back to where it came from. That traceability is not incidental. It is the feature that makes simulator practice useful rather than arbitrary.

Here is a concrete drill to build the source-vetting reflex:

Open any Speed Run session and, before responding to each scenario prompt, ask yourself: if this number were presented to me outside this simulator — in a screenshot, in a post, in an article — could I trace it back to its origin? Then, after the session, open the replay debrief. Observe how the simulator shows you the outcome data it used to score your decision. Ask: what would I have needed to know, at the time of the decision, to recognize whether the information driving that decision was primary or hearsay?

The goal is not to memorize historical prices. The goal is to develop a habit of asking "where does this number come from and how would I check it?" until that question becomes automatic before any piece of information influences your thinking.

Common Mistakes

  • Mistaking confidence for credibility. A source that speaks with certainty has not necessarily earned that certainty. Tone is not evidence tier.
  • Outsourcing the incentive check. Assuming that because a source is popular or widely followed it has already been vetted by others. Popularity and reliability are independent variables.
  • Stopping at "it's probably right." Probability of accuracy and verifiability are not the same thing. A claim might be correct for the wrong reasons, which means you cannot learn from it reliably even when it works out.
  • Applying vetting selectively. Scrutinizing sources that contradict an existing view while accepting sources that confirm it at face value. This is confirmation bias operating through the source-selection layer rather than the conclusion layer.
  • Treating recency as a quality signal. A very recent piece of information is not more reliable because it is recent. Its tier and incentive structure are unchanged by its timestamp.

Reflection Prompt

After your next simulator session, write down the last three pieces of information you recall encountering outside the simulator — a post, an article, a comment, anything market-related. For each one, answer: what tier is this source, what is their incentive, and could you verify the core claim from a primary source? Notice whether your first instinct was to evaluate the claim or the source. The instinct itself is data about where your process is currently developing.

Three-Question Check

  1. You see a post showing a large percentage gain on a trade, with no entry context, no strategy explanation, and an anonymous author. Which evidence tier does this belong to, and what is the single most important thing missing from the information?
  2. An analyst publishes a detailed breakdown of a company's earnings that accurately cites the official filing. This is Tier 2 — secondary commentary. What question should you ask about the analyst's incentives before treating their interpretation as neutral?
  3. You are reviewing a historical scenario in Abu's Speed Run. The outcome data shown comes from verified exchange records. How does the traceability of that data change how you should weight it compared to an anonymous screenshot showing the same percentage move?

Closing

Source hygiene is not a filter you apply to bad information. It is a discipline you apply to all information, before the claim inside it gets anywhere near a decision. The market does not reward you for acting on confident information. It is indifferent to confidence. What matters is whether the information you acted on was actually connected to reality in a traceable, verifiable way — and whether you understood the incentives of the person who handed it to you. Developing that discipline in a simulator, where the data is transparent and the outcomes are verifiable, is precisely the kind of practice that transfers. The habit of asking "where did this come from, and what does the source want?" costs nothing in a drill. In a real decision, it can be the difference between a reasoned process and someone else's agenda dressed up as your own conviction.

Educational simulator content, not financial advice.